File size: 24,113 Bytes
5ef0bb3 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 |
2023-10-17 09:40:41,085 ----------------------------------------------------------------------------------------------------
2023-10-17 09:40:41,086 Model: "SequenceTagger(
(embeddings): TransformerWordEmbeddings(
(model): ElectraModel(
(embeddings): ElectraEmbeddings(
(word_embeddings): Embedding(32001, 768)
(position_embeddings): Embedding(512, 768)
(token_type_embeddings): Embedding(2, 768)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(encoder): ElectraEncoder(
(layer): ModuleList(
(0-11): 12 x ElectraLayer(
(attention): ElectraAttention(
(self): ElectraSelfAttention(
(query): Linear(in_features=768, out_features=768, bias=True)
(key): Linear(in_features=768, out_features=768, bias=True)
(value): Linear(in_features=768, out_features=768, bias=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(output): ElectraSelfOutput(
(dense): Linear(in_features=768, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(intermediate): ElectraIntermediate(
(dense): Linear(in_features=768, out_features=3072, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): ElectraOutput(
(dense): Linear(in_features=3072, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=768, out_features=25, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-17 09:40:41,086 ----------------------------------------------------------------------------------------------------
2023-10-17 09:40:41,086 MultiCorpus: 1214 train + 266 dev + 251 test sentences
- NER_HIPE_2022 Corpus: 1214 train + 266 dev + 251 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/ajmc/en/with_doc_seperator
2023-10-17 09:40:41,086 ----------------------------------------------------------------------------------------------------
2023-10-17 09:40:41,086 Train: 1214 sentences
2023-10-17 09:40:41,086 (train_with_dev=False, train_with_test=False)
2023-10-17 09:40:41,086 ----------------------------------------------------------------------------------------------------
2023-10-17 09:40:41,086 Training Params:
2023-10-17 09:40:41,086 - learning_rate: "3e-05"
2023-10-17 09:40:41,086 - mini_batch_size: "4"
2023-10-17 09:40:41,087 - max_epochs: "10"
2023-10-17 09:40:41,087 - shuffle: "True"
2023-10-17 09:40:41,087 ----------------------------------------------------------------------------------------------------
2023-10-17 09:40:41,087 Plugins:
2023-10-17 09:40:41,087 - TensorboardLogger
2023-10-17 09:40:41,087 - LinearScheduler | warmup_fraction: '0.1'
2023-10-17 09:40:41,087 ----------------------------------------------------------------------------------------------------
2023-10-17 09:40:41,087 Final evaluation on model from best epoch (best-model.pt)
2023-10-17 09:40:41,087 - metric: "('micro avg', 'f1-score')"
2023-10-17 09:40:41,087 ----------------------------------------------------------------------------------------------------
2023-10-17 09:40:41,087 Computation:
2023-10-17 09:40:41,087 - compute on device: cuda:0
2023-10-17 09:40:41,087 - embedding storage: none
2023-10-17 09:40:41,087 ----------------------------------------------------------------------------------------------------
2023-10-17 09:40:41,087 Model training base path: "hmbench-ajmc/en-hmteams/teams-base-historic-multilingual-discriminator-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-2"
2023-10-17 09:40:41,087 ----------------------------------------------------------------------------------------------------
2023-10-17 09:40:41,087 ----------------------------------------------------------------------------------------------------
2023-10-17 09:40:41,087 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-17 09:40:42,550 epoch 1 - iter 30/304 - loss 4.00925954 - time (sec): 1.46 - samples/sec: 2025.07 - lr: 0.000003 - momentum: 0.000000
2023-10-17 09:40:44,107 epoch 1 - iter 60/304 - loss 3.29571244 - time (sec): 3.02 - samples/sec: 2034.37 - lr: 0.000006 - momentum: 0.000000
2023-10-17 09:40:45,575 epoch 1 - iter 90/304 - loss 2.50210834 - time (sec): 4.49 - samples/sec: 2078.76 - lr: 0.000009 - momentum: 0.000000
2023-10-17 09:40:46,958 epoch 1 - iter 120/304 - loss 2.06250957 - time (sec): 5.87 - samples/sec: 2083.13 - lr: 0.000012 - momentum: 0.000000
2023-10-17 09:40:48,324 epoch 1 - iter 150/304 - loss 1.74686239 - time (sec): 7.24 - samples/sec: 2147.08 - lr: 0.000015 - momentum: 0.000000
2023-10-17 09:40:49,607 epoch 1 - iter 180/304 - loss 1.53869916 - time (sec): 8.52 - samples/sec: 2147.16 - lr: 0.000018 - momentum: 0.000000
2023-10-17 09:40:50,903 epoch 1 - iter 210/304 - loss 1.37891942 - time (sec): 9.81 - samples/sec: 2169.99 - lr: 0.000021 - momentum: 0.000000
2023-10-17 09:40:52,181 epoch 1 - iter 240/304 - loss 1.24625376 - time (sec): 11.09 - samples/sec: 2195.01 - lr: 0.000024 - momentum: 0.000000
2023-10-17 09:40:53,524 epoch 1 - iter 270/304 - loss 1.13828827 - time (sec): 12.44 - samples/sec: 2212.10 - lr: 0.000027 - momentum: 0.000000
2023-10-17 09:40:54,853 epoch 1 - iter 300/304 - loss 1.04580906 - time (sec): 13.76 - samples/sec: 2231.07 - lr: 0.000030 - momentum: 0.000000
2023-10-17 09:40:55,031 ----------------------------------------------------------------------------------------------------
2023-10-17 09:40:55,031 EPOCH 1 done: loss 1.0377 - lr: 0.000030
2023-10-17 09:40:56,017 DEV : loss 0.196847066283226 - f1-score (micro avg) 0.5828
2023-10-17 09:40:56,027 saving best model
2023-10-17 09:40:56,403 ----------------------------------------------------------------------------------------------------
2023-10-17 09:40:57,759 epoch 2 - iter 30/304 - loss 0.22973336 - time (sec): 1.35 - samples/sec: 2261.94 - lr: 0.000030 - momentum: 0.000000
2023-10-17 09:40:59,251 epoch 2 - iter 60/304 - loss 0.19011469 - time (sec): 2.85 - samples/sec: 2194.87 - lr: 0.000029 - momentum: 0.000000
2023-10-17 09:41:00,618 epoch 2 - iter 90/304 - loss 0.18365433 - time (sec): 4.21 - samples/sec: 2166.12 - lr: 0.000029 - momentum: 0.000000
2023-10-17 09:41:02,006 epoch 2 - iter 120/304 - loss 0.17906680 - time (sec): 5.60 - samples/sec: 2167.64 - lr: 0.000029 - momentum: 0.000000
2023-10-17 09:41:03,473 epoch 2 - iter 150/304 - loss 0.16917255 - time (sec): 7.07 - samples/sec: 2164.28 - lr: 0.000028 - momentum: 0.000000
2023-10-17 09:41:04,821 epoch 2 - iter 180/304 - loss 0.16214454 - time (sec): 8.42 - samples/sec: 2170.92 - lr: 0.000028 - momentum: 0.000000
2023-10-17 09:41:06,286 epoch 2 - iter 210/304 - loss 0.15246777 - time (sec): 9.88 - samples/sec: 2153.09 - lr: 0.000028 - momentum: 0.000000
2023-10-17 09:41:07,746 epoch 2 - iter 240/304 - loss 0.14813300 - time (sec): 11.34 - samples/sec: 2151.10 - lr: 0.000027 - momentum: 0.000000
2023-10-17 09:41:09,180 epoch 2 - iter 270/304 - loss 0.14943578 - time (sec): 12.78 - samples/sec: 2162.70 - lr: 0.000027 - momentum: 0.000000
2023-10-17 09:41:10,552 epoch 2 - iter 300/304 - loss 0.14825159 - time (sec): 14.15 - samples/sec: 2170.02 - lr: 0.000027 - momentum: 0.000000
2023-10-17 09:41:10,730 ----------------------------------------------------------------------------------------------------
2023-10-17 09:41:10,730 EPOCH 2 done: loss 0.1473 - lr: 0.000027
2023-10-17 09:41:11,707 DEV : loss 0.15535008907318115 - f1-score (micro avg) 0.7654
2023-10-17 09:41:11,716 saving best model
2023-10-17 09:41:12,173 ----------------------------------------------------------------------------------------------------
2023-10-17 09:41:13,559 epoch 3 - iter 30/304 - loss 0.09876950 - time (sec): 1.38 - samples/sec: 2285.48 - lr: 0.000026 - momentum: 0.000000
2023-10-17 09:41:14,972 epoch 3 - iter 60/304 - loss 0.08919063 - time (sec): 2.80 - samples/sec: 2147.48 - lr: 0.000026 - momentum: 0.000000
2023-10-17 09:41:16,416 epoch 3 - iter 90/304 - loss 0.08479748 - time (sec): 4.24 - samples/sec: 2115.18 - lr: 0.000026 - momentum: 0.000000
2023-10-17 09:41:17,786 epoch 3 - iter 120/304 - loss 0.08416981 - time (sec): 5.61 - samples/sec: 2126.93 - lr: 0.000025 - momentum: 0.000000
2023-10-17 09:41:19,251 epoch 3 - iter 150/304 - loss 0.08841609 - time (sec): 7.07 - samples/sec: 2114.10 - lr: 0.000025 - momentum: 0.000000
2023-10-17 09:41:20,603 epoch 3 - iter 180/304 - loss 0.08959002 - time (sec): 8.43 - samples/sec: 2140.67 - lr: 0.000025 - momentum: 0.000000
2023-10-17 09:41:21,969 epoch 3 - iter 210/304 - loss 0.08360212 - time (sec): 9.79 - samples/sec: 2164.05 - lr: 0.000024 - momentum: 0.000000
2023-10-17 09:41:23,364 epoch 3 - iter 240/304 - loss 0.08313582 - time (sec): 11.19 - samples/sec: 2200.52 - lr: 0.000024 - momentum: 0.000000
2023-10-17 09:41:24,716 epoch 3 - iter 270/304 - loss 0.07903175 - time (sec): 12.54 - samples/sec: 2212.99 - lr: 0.000024 - momentum: 0.000000
2023-10-17 09:41:26,117 epoch 3 - iter 300/304 - loss 0.08135561 - time (sec): 13.94 - samples/sec: 2202.04 - lr: 0.000023 - momentum: 0.000000
2023-10-17 09:41:26,295 ----------------------------------------------------------------------------------------------------
2023-10-17 09:41:26,295 EPOCH 3 done: loss 0.0825 - lr: 0.000023
2023-10-17 09:41:27,316 DEV : loss 0.15102672576904297 - f1-score (micro avg) 0.8162
2023-10-17 09:41:27,326 saving best model
2023-10-17 09:41:27,813 ----------------------------------------------------------------------------------------------------
2023-10-17 09:41:29,243 epoch 4 - iter 30/304 - loss 0.03313024 - time (sec): 1.43 - samples/sec: 1954.02 - lr: 0.000023 - momentum: 0.000000
2023-10-17 09:41:30,597 epoch 4 - iter 60/304 - loss 0.05504111 - time (sec): 2.78 - samples/sec: 1993.06 - lr: 0.000023 - momentum: 0.000000
2023-10-17 09:41:31,945 epoch 4 - iter 90/304 - loss 0.06404305 - time (sec): 4.13 - samples/sec: 2039.54 - lr: 0.000022 - momentum: 0.000000
2023-10-17 09:41:33,306 epoch 4 - iter 120/304 - loss 0.05743194 - time (sec): 5.49 - samples/sec: 2067.00 - lr: 0.000022 - momentum: 0.000000
2023-10-17 09:41:34,696 epoch 4 - iter 150/304 - loss 0.05944006 - time (sec): 6.88 - samples/sec: 2108.15 - lr: 0.000022 - momentum: 0.000000
2023-10-17 09:41:36,115 epoch 4 - iter 180/304 - loss 0.06091412 - time (sec): 8.30 - samples/sec: 2143.11 - lr: 0.000021 - momentum: 0.000000
2023-10-17 09:41:37,453 epoch 4 - iter 210/304 - loss 0.06464068 - time (sec): 9.64 - samples/sec: 2167.63 - lr: 0.000021 - momentum: 0.000000
2023-10-17 09:41:38,782 epoch 4 - iter 240/304 - loss 0.06497940 - time (sec): 10.97 - samples/sec: 2192.80 - lr: 0.000021 - momentum: 0.000000
2023-10-17 09:41:40,119 epoch 4 - iter 270/304 - loss 0.06477435 - time (sec): 12.30 - samples/sec: 2218.30 - lr: 0.000020 - momentum: 0.000000
2023-10-17 09:41:41,492 epoch 4 - iter 300/304 - loss 0.06200751 - time (sec): 13.68 - samples/sec: 2236.85 - lr: 0.000020 - momentum: 0.000000
2023-10-17 09:41:41,687 ----------------------------------------------------------------------------------------------------
2023-10-17 09:41:41,687 EPOCH 4 done: loss 0.0626 - lr: 0.000020
2023-10-17 09:41:42,641 DEV : loss 0.1982080489397049 - f1-score (micro avg) 0.853
2023-10-17 09:41:42,649 saving best model
2023-10-17 09:41:43,137 ----------------------------------------------------------------------------------------------------
2023-10-17 09:41:44,473 epoch 5 - iter 30/304 - loss 0.02592012 - time (sec): 1.33 - samples/sec: 2385.92 - lr: 0.000020 - momentum: 0.000000
2023-10-17 09:41:45,801 epoch 5 - iter 60/304 - loss 0.04868729 - time (sec): 2.66 - samples/sec: 2231.74 - lr: 0.000019 - momentum: 0.000000
2023-10-17 09:41:47,152 epoch 5 - iter 90/304 - loss 0.04874142 - time (sec): 4.01 - samples/sec: 2241.31 - lr: 0.000019 - momentum: 0.000000
2023-10-17 09:41:48,527 epoch 5 - iter 120/304 - loss 0.05426644 - time (sec): 5.38 - samples/sec: 2203.02 - lr: 0.000019 - momentum: 0.000000
2023-10-17 09:41:49,849 epoch 5 - iter 150/304 - loss 0.05072023 - time (sec): 6.71 - samples/sec: 2193.33 - lr: 0.000018 - momentum: 0.000000
2023-10-17 09:41:51,190 epoch 5 - iter 180/304 - loss 0.04758743 - time (sec): 8.05 - samples/sec: 2263.13 - lr: 0.000018 - momentum: 0.000000
2023-10-17 09:41:52,577 epoch 5 - iter 210/304 - loss 0.04301930 - time (sec): 9.43 - samples/sec: 2282.60 - lr: 0.000018 - momentum: 0.000000
2023-10-17 09:41:53,927 epoch 5 - iter 240/304 - loss 0.04057396 - time (sec): 10.78 - samples/sec: 2274.76 - lr: 0.000017 - momentum: 0.000000
2023-10-17 09:41:55,260 epoch 5 - iter 270/304 - loss 0.04417344 - time (sec): 12.12 - samples/sec: 2266.18 - lr: 0.000017 - momentum: 0.000000
2023-10-17 09:41:56,631 epoch 5 - iter 300/304 - loss 0.04586453 - time (sec): 13.49 - samples/sec: 2269.48 - lr: 0.000017 - momentum: 0.000000
2023-10-17 09:41:56,810 ----------------------------------------------------------------------------------------------------
2023-10-17 09:41:56,810 EPOCH 5 done: loss 0.0458 - lr: 0.000017
2023-10-17 09:41:57,802 DEV : loss 0.19336670637130737 - f1-score (micro avg) 0.8592
2023-10-17 09:41:57,811 saving best model
2023-10-17 09:41:58,356 ----------------------------------------------------------------------------------------------------
2023-10-17 09:41:59,948 epoch 6 - iter 30/304 - loss 0.01221848 - time (sec): 1.59 - samples/sec: 2075.28 - lr: 0.000016 - momentum: 0.000000
2023-10-17 09:42:01,580 epoch 6 - iter 60/304 - loss 0.02836975 - time (sec): 3.22 - samples/sec: 2062.77 - lr: 0.000016 - momentum: 0.000000
2023-10-17 09:42:03,169 epoch 6 - iter 90/304 - loss 0.02441422 - time (sec): 4.81 - samples/sec: 2022.26 - lr: 0.000016 - momentum: 0.000000
2023-10-17 09:42:04,789 epoch 6 - iter 120/304 - loss 0.01997158 - time (sec): 6.43 - samples/sec: 1989.89 - lr: 0.000015 - momentum: 0.000000
2023-10-17 09:42:06,367 epoch 6 - iter 150/304 - loss 0.02226174 - time (sec): 8.01 - samples/sec: 1976.24 - lr: 0.000015 - momentum: 0.000000
2023-10-17 09:42:07,771 epoch 6 - iter 180/304 - loss 0.03289946 - time (sec): 9.41 - samples/sec: 1970.20 - lr: 0.000015 - momentum: 0.000000
2023-10-17 09:42:09,129 epoch 6 - iter 210/304 - loss 0.03414661 - time (sec): 10.77 - samples/sec: 1983.66 - lr: 0.000014 - momentum: 0.000000
2023-10-17 09:42:10,498 epoch 6 - iter 240/304 - loss 0.03480877 - time (sec): 12.14 - samples/sec: 2008.11 - lr: 0.000014 - momentum: 0.000000
2023-10-17 09:42:11,866 epoch 6 - iter 270/304 - loss 0.03370047 - time (sec): 13.51 - samples/sec: 2030.09 - lr: 0.000014 - momentum: 0.000000
2023-10-17 09:42:13,213 epoch 6 - iter 300/304 - loss 0.03260773 - time (sec): 14.85 - samples/sec: 2062.93 - lr: 0.000013 - momentum: 0.000000
2023-10-17 09:42:13,393 ----------------------------------------------------------------------------------------------------
2023-10-17 09:42:13,393 EPOCH 6 done: loss 0.0323 - lr: 0.000013
2023-10-17 09:42:14,350 DEV : loss 0.20636732876300812 - f1-score (micro avg) 0.864
2023-10-17 09:42:14,363 saving best model
2023-10-17 09:42:14,889 ----------------------------------------------------------------------------------------------------
2023-10-17 09:42:16,291 epoch 7 - iter 30/304 - loss 0.03710421 - time (sec): 1.39 - samples/sec: 2059.99 - lr: 0.000013 - momentum: 0.000000
2023-10-17 09:42:17,714 epoch 7 - iter 60/304 - loss 0.03619170 - time (sec): 2.81 - samples/sec: 2069.57 - lr: 0.000013 - momentum: 0.000000
2023-10-17 09:42:19,250 epoch 7 - iter 90/304 - loss 0.03524165 - time (sec): 4.35 - samples/sec: 2081.46 - lr: 0.000012 - momentum: 0.000000
2023-10-17 09:42:20,676 epoch 7 - iter 120/304 - loss 0.03009571 - time (sec): 5.77 - samples/sec: 2080.39 - lr: 0.000012 - momentum: 0.000000
2023-10-17 09:42:22,138 epoch 7 - iter 150/304 - loss 0.02823354 - time (sec): 7.24 - samples/sec: 2083.35 - lr: 0.000012 - momentum: 0.000000
2023-10-17 09:42:23,580 epoch 7 - iter 180/304 - loss 0.02821860 - time (sec): 8.68 - samples/sec: 2057.68 - lr: 0.000011 - momentum: 0.000000
2023-10-17 09:42:25,034 epoch 7 - iter 210/304 - loss 0.02646873 - time (sec): 10.13 - samples/sec: 2064.79 - lr: 0.000011 - momentum: 0.000000
2023-10-17 09:42:26,453 epoch 7 - iter 240/304 - loss 0.02543135 - time (sec): 11.55 - samples/sec: 2084.44 - lr: 0.000011 - momentum: 0.000000
2023-10-17 09:42:27,915 epoch 7 - iter 270/304 - loss 0.02847803 - time (sec): 13.01 - samples/sec: 2099.39 - lr: 0.000010 - momentum: 0.000000
2023-10-17 09:42:29,300 epoch 7 - iter 300/304 - loss 0.02760448 - time (sec): 14.40 - samples/sec: 2129.01 - lr: 0.000010 - momentum: 0.000000
2023-10-17 09:42:29,493 ----------------------------------------------------------------------------------------------------
2023-10-17 09:42:29,494 EPOCH 7 done: loss 0.0273 - lr: 0.000010
2023-10-17 09:42:30,447 DEV : loss 0.20442555844783783 - f1-score (micro avg) 0.844
2023-10-17 09:42:30,454 ----------------------------------------------------------------------------------------------------
2023-10-17 09:42:31,792 epoch 8 - iter 30/304 - loss 0.03331394 - time (sec): 1.34 - samples/sec: 2206.88 - lr: 0.000010 - momentum: 0.000000
2023-10-17 09:42:33,128 epoch 8 - iter 60/304 - loss 0.01684349 - time (sec): 2.67 - samples/sec: 2228.46 - lr: 0.000009 - momentum: 0.000000
2023-10-17 09:42:34,480 epoch 8 - iter 90/304 - loss 0.01445152 - time (sec): 4.02 - samples/sec: 2291.00 - lr: 0.000009 - momentum: 0.000000
2023-10-17 09:42:35,829 epoch 8 - iter 120/304 - loss 0.01339351 - time (sec): 5.37 - samples/sec: 2256.72 - lr: 0.000009 - momentum: 0.000000
2023-10-17 09:42:37,179 epoch 8 - iter 150/304 - loss 0.02202332 - time (sec): 6.72 - samples/sec: 2246.82 - lr: 0.000008 - momentum: 0.000000
2023-10-17 09:42:38,543 epoch 8 - iter 180/304 - loss 0.01918384 - time (sec): 8.09 - samples/sec: 2269.07 - lr: 0.000008 - momentum: 0.000000
2023-10-17 09:42:39,869 epoch 8 - iter 210/304 - loss 0.01845123 - time (sec): 9.41 - samples/sec: 2257.43 - lr: 0.000008 - momentum: 0.000000
2023-10-17 09:42:41,238 epoch 8 - iter 240/304 - loss 0.02065770 - time (sec): 10.78 - samples/sec: 2282.79 - lr: 0.000007 - momentum: 0.000000
2023-10-17 09:42:42,601 epoch 8 - iter 270/304 - loss 0.02081829 - time (sec): 12.15 - samples/sec: 2274.37 - lr: 0.000007 - momentum: 0.000000
2023-10-17 09:42:43,940 epoch 8 - iter 300/304 - loss 0.02159129 - time (sec): 13.49 - samples/sec: 2272.19 - lr: 0.000007 - momentum: 0.000000
2023-10-17 09:42:44,118 ----------------------------------------------------------------------------------------------------
2023-10-17 09:42:44,118 EPOCH 8 done: loss 0.0218 - lr: 0.000007
2023-10-17 09:42:45,142 DEV : loss 0.21142761409282684 - f1-score (micro avg) 0.8647
2023-10-17 09:42:45,150 saving best model
2023-10-17 09:42:45,635 ----------------------------------------------------------------------------------------------------
2023-10-17 09:42:47,017 epoch 9 - iter 30/304 - loss 0.00896017 - time (sec): 1.38 - samples/sec: 2060.55 - lr: 0.000006 - momentum: 0.000000
2023-10-17 09:42:48,418 epoch 9 - iter 60/304 - loss 0.01136709 - time (sec): 2.78 - samples/sec: 2136.14 - lr: 0.000006 - momentum: 0.000000
2023-10-17 09:42:49,776 epoch 9 - iter 90/304 - loss 0.01848839 - time (sec): 4.14 - samples/sec: 2168.92 - lr: 0.000006 - momentum: 0.000000
2023-10-17 09:42:51,111 epoch 9 - iter 120/304 - loss 0.01553642 - time (sec): 5.47 - samples/sec: 2237.45 - lr: 0.000005 - momentum: 0.000000
2023-10-17 09:42:52,448 epoch 9 - iter 150/304 - loss 0.01360062 - time (sec): 6.81 - samples/sec: 2241.17 - lr: 0.000005 - momentum: 0.000000
2023-10-17 09:42:53,795 epoch 9 - iter 180/304 - loss 0.01608852 - time (sec): 8.16 - samples/sec: 2233.62 - lr: 0.000005 - momentum: 0.000000
2023-10-17 09:42:55,125 epoch 9 - iter 210/304 - loss 0.01555607 - time (sec): 9.49 - samples/sec: 2219.00 - lr: 0.000004 - momentum: 0.000000
2023-10-17 09:42:56,493 epoch 9 - iter 240/304 - loss 0.01405601 - time (sec): 10.86 - samples/sec: 2208.17 - lr: 0.000004 - momentum: 0.000000
2023-10-17 09:42:57,909 epoch 9 - iter 270/304 - loss 0.01755853 - time (sec): 12.27 - samples/sec: 2235.31 - lr: 0.000004 - momentum: 0.000000
2023-10-17 09:42:59,338 epoch 9 - iter 300/304 - loss 0.01701838 - time (sec): 13.70 - samples/sec: 2226.40 - lr: 0.000003 - momentum: 0.000000
2023-10-17 09:42:59,518 ----------------------------------------------------------------------------------------------------
2023-10-17 09:42:59,519 EPOCH 9 done: loss 0.0171 - lr: 0.000003
2023-10-17 09:43:00,556 DEV : loss 0.21606358885765076 - f1-score (micro avg) 0.8657
2023-10-17 09:43:00,564 saving best model
2023-10-17 09:43:01,075 ----------------------------------------------------------------------------------------------------
2023-10-17 09:43:02,399 epoch 10 - iter 30/304 - loss 0.01079333 - time (sec): 1.32 - samples/sec: 2282.69 - lr: 0.000003 - momentum: 0.000000
2023-10-17 09:43:03,827 epoch 10 - iter 60/304 - loss 0.00535025 - time (sec): 2.75 - samples/sec: 2240.61 - lr: 0.000003 - momentum: 0.000000
2023-10-17 09:43:05,182 epoch 10 - iter 90/304 - loss 0.01128766 - time (sec): 4.10 - samples/sec: 2280.98 - lr: 0.000002 - momentum: 0.000000
2023-10-17 09:43:06,536 epoch 10 - iter 120/304 - loss 0.01589802 - time (sec): 5.46 - samples/sec: 2236.46 - lr: 0.000002 - momentum: 0.000000
2023-10-17 09:43:07,903 epoch 10 - iter 150/304 - loss 0.01443741 - time (sec): 6.82 - samples/sec: 2250.57 - lr: 0.000002 - momentum: 0.000000
2023-10-17 09:43:09,317 epoch 10 - iter 180/304 - loss 0.01637260 - time (sec): 8.24 - samples/sec: 2213.08 - lr: 0.000001 - momentum: 0.000000
2023-10-17 09:43:10,694 epoch 10 - iter 210/304 - loss 0.01591658 - time (sec): 9.62 - samples/sec: 2216.92 - lr: 0.000001 - momentum: 0.000000
2023-10-17 09:43:12,129 epoch 10 - iter 240/304 - loss 0.01523204 - time (sec): 11.05 - samples/sec: 2219.38 - lr: 0.000001 - momentum: 0.000000
2023-10-17 09:43:13,453 epoch 10 - iter 270/304 - loss 0.01368850 - time (sec): 12.37 - samples/sec: 2208.39 - lr: 0.000000 - momentum: 0.000000
2023-10-17 09:43:14,832 epoch 10 - iter 300/304 - loss 0.01292642 - time (sec): 13.75 - samples/sec: 2223.75 - lr: 0.000000 - momentum: 0.000000
2023-10-17 09:43:15,004 ----------------------------------------------------------------------------------------------------
2023-10-17 09:43:15,004 EPOCH 10 done: loss 0.0128 - lr: 0.000000
2023-10-17 09:43:16,004 DEV : loss 0.21460825204849243 - f1-score (micro avg) 0.8647
2023-10-17 09:43:16,378 ----------------------------------------------------------------------------------------------------
2023-10-17 09:43:16,379 Loading model from best epoch ...
2023-10-17 09:43:17,813 SequenceTagger predicts: Dictionary with 25 tags: O, S-scope, B-scope, E-scope, I-scope, S-pers, B-pers, E-pers, I-pers, S-work, B-work, E-work, I-work, S-loc, B-loc, E-loc, I-loc, S-date, B-date, E-date, I-date, S-object, B-object, E-object, I-object
2023-10-17 09:43:18,725
Results:
- F-score (micro) 0.8263
- F-score (macro) 0.6655
- Accuracy 0.7091
By class:
precision recall f1-score support
scope 0.7756 0.8013 0.7883 151
work 0.7757 0.8737 0.8218 95
pers 0.9082 0.9271 0.9175 96
date 0.0000 0.0000 0.0000 3
loc 1.0000 0.6667 0.8000 3
micro avg 0.8060 0.8477 0.8263 348
macro avg 0.6919 0.6538 0.6655 348
weighted avg 0.8075 0.8477 0.8264 348
2023-10-17 09:43:18,725 ----------------------------------------------------------------------------------------------------
|